effect size

This forum is closed, and read-only.
MarketingStudent
PLS Junior User
Posts: 5
Joined: Tue Aug 03, 2010 9:53 pm
Real name and title:

Effect size or not;Cohen's formula not suitable for my model

Post by MarketingStudent »

Dear all, dear Mr. Nitzl,

you wrote that it might not be necessary to report the f² for Reem's model and that R² is enough.

I am also not sure, if the calcualtion of the f² in my model makes sense.It's similar, but a bit different from the models described above.

A --> B--> C
B--> D
B--> E

The relation I am interested to test is mainly between B (my main construct) and the three outcome variables (C,D, and E)

However with the formula as suggested by Cohen (1988) this does not work, since I cannot leave B out of the model (david all the way on the top describes this problem). Professor Bido, however, suggested an approach, namely reporting the correlation between B and C (see below point 3) ), being equivalent to the effect size. However I don't know if I can apply this, since I have more than only C as dependent variable, therefore different to the model Bido was refering to.

I was mainly interested in reporting the effect size, since the three path coefficients each between B and the the dependent variables (C,D, E) turned out to be very low (0.11), however, the R squares of C, D, E reach the minimal acceptable level of 0.19. Therefore, I thought that by calcualting the effect size of B on each of the outcome variables I could demonstrate that hte path coefficients are still valuable, even though their so low, and that B has actually an effect (may be weak but still) on the outcome variables.

Does that make sense?

[ad 1) In my opinion R2 by itself is enough. Because R2 is the percentage of the explained variance of your focal variable C through your model. In other words your whole “effect” . Why do you need the effect size for your model if you won’t test the separate effects of the mediating variables?]
Diogenes wrote:Hi David,

I looked for more information, and…

1) f2 = R2 / (1 - R2) --> used in multiple regression considering all independent variables (It is not what you are trying to assess).

2) (R2included - R2 excluded)/(1 - R2 included) --> used in multiple regression considering the partial coefficients (ok for A and B variables, ok as you have done).

3) Eventually, as you have just one relation ( C --> D ) the beta = r, then, the effect size will be the value of this correlation (Cohen, 1977, p.75-80), with:
r = 0.10 = small effect
r = 0.30 = medium effect
r = 0.50 = large effect

I hope this help.

Bido
Locked